HD44780A00 LCD - half of it doesn't work

>>1137686
You need to initialize it.

>>1137741

You mean without initialization it's normal that it displays only half of the line...? I found a post here:

http://forum.arduino.cc/index.php?topic=111894.0

...where a guy says: "Next you should connect the power and contrast connections (pins 1, 2, and 3). You should be able to get a display consisting of one line of 'blocks'." In my case, I got only half of block lit... And wonder if there's some kind of hardware problem with the paths on PCB.

>>1137686
stop being negative. Half of it works perfectly!

>>1137879

Well, I'm not so sure about that... If the backlight doesn't turn on and half of the blocks are dimmed maybe the communication part is also broken...? That's the reason for my question, I am no sure about how should I perform proper diagnostics on this thing.

Regards

The backlight is not tied directly to the power and not programmable, you need to power it separately.

The whole screen should have the same contrast, check your pinout. Then try pressing slightly on the sides of the screen, there might be a bad connection.

And draw a diagram of how you're connecting everything. Both for us and yourself.

File: scheme.jpg (77KB, 1026x598px) Image search: [Google]

77KB, 1026x598px

>>1137912
All righty. That's what I thought, that backlight should work no matter what (pins 15/16), but it doesn't. When I connect 5V/GND as shown on the attached scheme, the chip gets hot and that's it. Tried to apply pressure on the LCD as suggested but no dice :/ That's also what I thought, the whole light should lit up without a problem!

Ok, will give it a shot anyway, even though I'm not sure if it's operational... I have disassembled the LCD module - there was some dirt underneath which I removed using alcohol, but the problem with half-lit screen remains, unfortunately. Paths don't seem to be interrupted at any point. I am not an expert, but there's no backlight as far as I can tell - thus this begs a question what's the purpose of the additional pin pair...

Anyway, the chip doesn't get hot any more when I connect 5Vdc to those, as long as the polarization is correct (DC to pin 15, GND to pin 16)

Are you powering it through the MC? If so I'd hook it up to a separate power supply, you can usually jerry-rig an old USB charger to get plenty of amperage.

>>1138238
Started connecting everything the way it should be done, on a piece of plywood. Yeah, I'm using a separate power supply with 1A, so that's not an issue.

If anyone's following and is interested - The connector has 14 pins, not 16 - so there's no additional voltage pair for backlight. I have no clue how could I be so stupid, it would help to count those goddamn lines... What it means is that I was trying to connect 5VDC to data lines D6 and D7. Wonder if that could mess up something inside the chip.

I will continue with this LCD for science purposes. The rest of the screen is operational from what I can see, but just on the lowest possible contrast level. We will see if it will speak with LPT on my computer. Gonna update the post here when I'm finished.

>>1138358
>could mess up something inside the chip
Most definitely could have. Chips have a limit on how much they can source/sink.
While those pins should normally be input and high impedance, the chip getting hot would indicate that they were sinking/sourcing current.

And don't connect the contrast directly to ground/vcc. Use a pot connected to the three.

>LPT
You still have one of those on your PC? I'm kind of jealous. When I was messing around with 1602 screens I just wrote a driver for an atmega - they're not of much use connected to a computer anyway.

File: 17036048_395418094165577_763081026_o.jpg (89KB, 1280x720px) Image search: [Google]

89KB, 1280x720px

>>1138458
Thanks for the feedback! Now the good news is, I managed to hook it up to LPT - yep, still got one of those, together winth Windows 98 - for diagnostics and other stuff which is impossible to perform on my T60... Anyways, the good news is, all blocks lit up after starting up a program for HD44780, so it's not damaged after all!

You're right about the potenciometer, I just don't have one ATM. Will try to play around with this thing tomorrow!

Remember folks, always count the pins! DOH...

>>1138483
I'd be lying if I said I never messed up the pins.

What are you going to use this for btw?

>>1138824

I am trying to learn some assembly basics, was trying to find a place to start and found this thing lying on the pavement :) Other than that, it would be nice to convert this baby into a simple POST counter for mobos.

I was unable to get anything more beside the picture I have posted before, even though I have checked all of my connections a dozen times. I have ordered LPT cable, sockets, diagnostic board, soldering iron plus some additional simple electronic components - the moment I got the package, I will make another attempt at this one.

>>1139264
>I was unable to get anything more beside the picture I have posted before
What did you try? From the pictures it doesn't look like you did anything more than just power it up.

Looks like you need to initialize it as a 2 line lcd.

I tried to connect the 8-line data bus and 3 additional ones, but got them wrong AFAIK - don't have a spare LPT plug, so it's a bit messy, don't want to play with it anymore until I get a board, a plug and some other stuff.

Anyway, even though it didn't display text after initialization, still all 16 blocks lit, so there's high chance it's working. I will trace all lines back to the chip on PCB, because it's possible this particular LCD doesn't follow regular schematics found over the Net (or there was a short circuit somewhere, I messed it up in short).